Although the genetic basis of complex diseases such as RA is likely to be multifactorial, the strong HLA-DR4 association with RA suggests that one of the contributing genetic features is encoded within HLA. Our previous work documents extensive structural and genotypic diversity of class II molecules among different normal DR4+ individuals. At least six different haplotypes, differing in gene products of at least three class II loci, all share the HLA-DR4 specificity, even though they are considered phenotypically identical by standard typing and disease association studies. It is now appropriate to reconsider the association between HLA-DR4 and RA in terms of a more comprehensive molecular context. In preliminary studies among DR4+ patients with seropositive JRA, a striking prevalence of two specific DR4+ haplotypes has been found. Using a panel of locus-specific monoclonal antibodies which distinguish the products of three separate class II loci on these two haplotypes we will evaluate molecular variants by 2-D gels, by peptide mapping, and by analysis of combinatorial associations for each of the different class II products linked to DR4. Substantial DNA variation within HLA class II genes also is observed among DR4+ haplotypes; to examine whether any such polymorphisms represent consistent predictive markers for DR4+ haplotypes in RA, we will extend these studies to JRA and adult RA populations analyzed both by RFLP with DR, DQ, and DP cDNA probes, and by B1 exon sequencing techniques. These studies will yield a molecular fingerprint, both at the genomic and the polypeptide level, for a series of class II molecules present on DR4+ haplotypes in patients and normals. Such information is crucial for a more critical perspective on the issue of HLA and disease, and will also contribute to a number of basic clinical and immunogenetic questions currently unresolved.